Вы находитесь на странице: 1из 10

Pakistan Journal of Social Sciences (PJSS)

Vol. 35, No. 2 (2015), pp. 533-542

Role of Infrastructure in Poverty Alleviation:

Evidence from Pakistan
Muhammad Zahir Faridi, PhD
Associate Professor, School of Economics,
Bahauddin Zakariya University, Multan

Muhammad Omer Chaudhry, PhD

Assistant Professor, School of Economics,
Bahauddin Zakariya University, Multan

Muhammad Ramzan
Lecturer of Economics
Government Degree College, Mian Channu.

Poverty is one of the biggest issues confronting Pakistans economy
that is severely hampering the way of economic development and
growth. This paper investigates the role of infrastructure provision in
poverty alleviation in Pakistan. The present study employs a log-log
model in OLS settings to estimate the relationship between
infrastructure provision and poverty alleviation in Pakistan by using
time series data ranging 1975-2013. In order to find the relative impact
of infrastructure, the infrastructure is divided into two components i.e.
physical infrastructure and social infrastructure. The results indicate
that both social and physical infrastructures significantly reduce

Keywords: Social and physical infrastructure, Poverty, log- log model

I. Introduction
Poverty is a global issue but the severity of this problem is many folds in
developing countries like Pakistan. Poverty is one of the major obstacles in the process of
development and economic growth. Composite infrastructure (social and physical
infrastructure) plays an important role in the development of economy and reduction in
poverty. In recent years many studies have focused on composite effects of infrastructure
on economic growth but in this study we have estimated the relation between
infrastructure provision and poverty alleviation in Pakistan. This study has policy
implications as the provision of infrastructure has significant impact on poverty

Poverty has been defined in many ways and various forms of poverty are existing
in the literature such as lack of food, shelter, education and medication, lack of resources,
lack of opportunities and choices, violation of ones integrity, personal freedom etc.
Poverty also means powerlessness, exclusion and insecurity of life. World Bank defines
poverty as incapability to maintain a minimum living standard anticipated with respect
to basic consumption needs or some amount of income required for satisfying them
(World Bank 2006). In short, poverty is a multi dimensional phenomenon that
534 Pakistan Journal of Social Sciences Vol. 35, No. 2

encompasses every sphere of life. The problem and alleviation of poverty has been a
major policy focus in developing countries for a long time. It also takes central position
in the millennium development goals; the first of the eight goals is to half the amount of
the worlds residents living under 1.25 per day (United Nation, 2010). The extent of
poverty has been investigated by different methods by different scholars.

Infrastructure provision is considered very crucial in order to achieve desired

development and growth goals in an economy. Infrastructure is considered as capital
stock that is meant to provide public goods and services. Infrastructure provision has
direct impact on the production activities in an economy that eventually raises the living
standard and quality of life. There are numerous sources of infrastructure such as health
care, waste removal, education, transportation, telecommunication, electricity etc. The
impacts of public infrastructure include increase in productivity and economic activities,
reduction in transaction costs, increased employment opportunities and many more.
These all affects eventually reduce the poverty levels.

The present study aims to explore the role of infrastructure in poverty alleviation
in Pakistan. In order to establish the link between infrastructure provision and extent of
poverty, two infrastructure indices are estimated in this study i.e. physical infrastructure
index and social infrastructure index. The time period under consideration is 1975-2013.

II. State of Infrastructure in Pakistan

According to the World Competitive Report 2012-13 Pakistans physical stock is
less effective as compared to other developing countries in same income group. There is
a severe need to increase the ability of present infrastructure stock. Table 1 shows the
comparative ranking of Pakistan with respect to infrastructure in the region.

Table 1: Comparative ranking of Pakistan

Global Quality of Quality Quality of Quality of
Competitive Electricity of Railroads Transport
Index supply Roads
Pakistan 124 126 73 66 88
India 59 110 86 27 68
China 29 59 54 22 70
Iran 66 60 68 45 132
Nepal 125 143 127 122 129
Bangladesh 118 136 113 73 131
Source: Global Competitive Report 2012-13

Table 1 clearly shows the standing of Pakistan in the region with respect to quality
of infrastructure. The overall competitive index of Pakistan is the lowest among the
neighboring countries. The above statistics indicate that the infrastructure in Pakistan is
poorly managed and one of the results of this poor management is extreme energy crisis
during last few years.

A. Electricity
Pakistan is severely hit by the energy crisis during last decade. Table 2 illustrates
the supply and demand position of electricity in Pakistan.
Muhammad Zahir Faridi, Muhammad Omer Chaudhry, Muhammad Ramzan 535
Table 2: Supply-Demand Position of Electricity in Pakistan
Firm Supply(MW) Peak Demand (MW) Surplus / Deficit (MW)
2003 14,336 12,800 + 1,536
2004 15,046 13,831 + 1,215
2005 15,082 14,642 + 440
2006 15,072 15,483 - 411
2007 15,091 16,548 -1,457
2008 15,055 17,689 -2,634
2009 15,055 19,080 -4,025
2010 15,055 20,584 -5,529
Source: Pakistan Economic Survey (various issues)

Table 2 shows that the there is continuous and increasing deficit between supply
and demand of electricity in Pakistan during last decade.

B. Social Infrastructure
Pakistan has to spend most part of its budget on defense and debt servicing, thus a
small portion is left for development projects. Table 3 explains the R & D expenditure as
percentage of GDP in Pakistan

Table 3: Research and Development Expenditures as a percentage of GDP

R & D expenditure R & D expenditure
Years Years
(% of GDP) (% of GDP)
1997 0.16 2002 0.22
1998 0.11 2005 0.44
1999 0.12 2007 0.67
2000 0.13 2009 0.46
2001 0.17 2012 0.51
Source: Pakistan Economic Survey (Various issues)

It is evident from Table 3 that the share of R&D expenditures in Pakistan is very
low and there is no significant change over time. Table 4 show the expenditure on
education (percentage of GDP) in Pakistan.

Table 4: Education expenditure as Percentage of GDP

Education expenditure Education expenditure
Years Years
(% of GDP) (% of GDP)
1971 1.65 2004 1.95
1975 2.01 2005 2.25
1980 2.13 2007 2.84
1985 2.44 2008 2.91
1998 2.03 2009 2.69
1999 2.61 2010 2.37
2000 1.84 2011 2.23
2003 1.95 2012 2.17
Source: World Bank Indicators

Table 4 demonstrates the state of education expenditures in Pakistan. Being a

developing country, the provision of education is very important in order to achieve the
targets of higher economic growth and development. The current expenditure pattern on
education may not be enough to get rid of problem of illiteracy in Pakistan.
536 Pakistan Journal of Social Sciences Vol. 35, No. 2

The expenditures on health are even lower in Pakistan. Table 5 shows the
expenditure on health services in Pakistan.

Table 5: Health and Nutrition Expenditures in Pakistan

Health expenditure Health expenditure
Years Years
(% of GDP) (% of GDP)
2000-01 0.72 2007-08 0.57
2001-02 0.59 2008-09 0.56
2002-03 0.58 2009-10 0.54
2003-04 0.57 2010-11 0.23
2004-05 0.57 2011-12 0.27
2005-06 0.51 2012-13 0.35
2006-07 0.57
Source: Pakistan Economic Survey (Various issues)

III. Review of Literature

Poverty being a multidimensional phenomenon has been treated differently in
literature in its estimation. Studies have used form basic head count ration to the lack of
opportunities and personal freedom as indicators of poverty. However, regardless of
technique and indicator of poverty, almost every definition of poverty converges to the
deficiency to meet basic requirements of living.

There is growing literature on the links between infrastructure provision and

economic development but still such studies are missing in the context of Pakistan. The
debate on the linkages between infrastructure and economic growth was highlighted by
the work of Ascheur (1989) that showed a positive impact of public capital on economic
growth. However, the results of this study was subject to great criticism but still this is
regarded as one of t he pioneering studies in this field. Later on many studies tried to
estimate the link between infrastructure provision and economic growth in cross country
settings (see Sanchez-Robles 1998, Canning and Pedroni 2008, Devarajan et al 1996).

In case of Pakistan, there are not many studies that had worked on the relationship
between infrastructure and economic growth in general and poverty alleviation in
particular. Ahmed et al. (2013) has given a survey of the studies that have shown a
negative or insignificant impact of investment in infrastructure on economic growth.
Such studies include Ghani and Din (2006), Rehman et al (2010) and Planning
Commission (2011).

Faridi et al (2011) estimated the impacts of transportation and telecommunication

on economic growth in Pakistan. The authors used Solow growth model to estimate the
link by adopting ARDL approach on the data from 1972-2010. The authors suggested
that the government should have trained the labor technically to reduce poverty in
Pakistan. Similarly Imran and Niazi (2011) estimated the influence of infrastructure on
economic development in Pakistan. The study concluded that the slow pace of
development of infrastructure is the main cause of low economic growth. The results of
the study indicated that infrastructure have significant impact on total factor productivity
as well as on economic growth. The study suggested that a massive buildup of
infrastructure stock especially in electricity, telecommunication, transport and water
supply is essential in order to have positive significant impact on economic growth.
Muhammad Zahir Faridi, Muhammad Omer Chaudhry, Muhammad Ramzan 537
Ahmed et al (2013) pointed out the role of infrastructure in economic development
social welfare has been given special focus in literature of the past three decades. They
used CGE model interlinked with micro simulation model to highlight the overall impacts
of public infrastructure investment. They concluded that public infrastructure investment
by taxation or international loans have the same results or impacts. They viewed that tax
financing constrained output in industrial sector and reduce economic development in
short run and reduces export.

Sahoo et al (2010) conducted a study to analyze the role of infrastructure in

economic growth in China by developing a composite index of leading physical
infrastructure indicators. They used time series data for the period 1975 to 2007. The
results of the study showed that there were positive effects of domestic private
investment, domestic public investment, infrastructure index, total labor force and per
capita real public expenditures on health and education on gross domestic product.

Ogun (2010) investigated the role of infrastructural development on poverty

reduction in Nigeria, by using secondary data for the period 1970 to 2005 by employing
the structural vector autoregressive (SVAR) technique. The study pointed out that
infrastructural development played a vital role in poverty reduction. The study suggested
that gigantic investment in social infrastructure in cities would significantly reduce
poverty in the urban areas. Similarly, Jerome (2011) evaluated the impact of
infrastructure and economic growth with poverty alleviation in Africa and found it to be

IV. Data, Variables and Methodology

The empirical analysis based on quantitative data and appropriate estimation
technique, the data source and suitable methodology is explained below.

A. Data Sources
The present analysis is based on secondary source of data. The time series data is
used in the analysis considering the time period 1975 to 2013. Data is generated from
different sources like World Development Indicators (WDI), various issues of Pakistan
Economic Survey, Handbook of Statistics and different published articles.

B. Methodology and Model Specification

In the present study we have applied the Ordinary Least Square (OLS) method for
multivariate data analysis. The general format of the multiple regression models is given

Yi= o + 1 X1 + 2 X2+.. +K XK + i

Where Yi is regressand variable, X1, X2, , Xk are explanatory variables where

1, 2,.., K are slope coefficients or partial regression coefficients. i is stochastic error
term that follows all OLS assumptions.

Considering the above technique, our specified function is given as:


538 Pakistan Journal of Social Sciences Vol. 35, No. 2

In order to calculate elasticities, we have used log-log form of the multiple

regression models:

LnPOV = o + 1LnLFP + 2LnFRTY + 3LnPI + 4LnSI


Where o is intercept term and 1, 2, 3, .. 7 are elasticities. The variable

description and their measurement issues are discussed briefly.

Poverty (POV)
The poverty is dependent variable. The present study uses Head Count Ratio as
measure of poverty.

Labor Force (LFP)

We have taken employed labor force in million people from the various issues of
Pakistan Economic Survey. Theoretically, it is hypothesized that there is inverse
relationship between poverty and employed labor force.

Fertility Rate (FRTY)

Fertility rate is considered a main indicator of poverty alleviation in developing
countries. According to microeconomic fertility theory, household demand more children
in order to raise their family income. Theoretically, it is expected that fertility reduces
poverty. Whereas, according to the supply side theories, the poverty level increases with
the increase in population.

It is core variable in the present study. It is generally assumed that provision of
infrastructure plays a vital role in reducing poverty. It has been observed that many
indicators have been used as proxy of infrastructure stock in different studies. These
include electricity, roads, railways, number of school, electricity, health expenditures etc.
This may create the problem of multicollinearity. In order to avoid the issue of
multicollinearity, we have constructed infrastructure index by employing principal
component analysis. In order to strengthen the analysis, we have divided the
infrastructure index into physical infrastructure index and social infrastructure index.

Physical Infrastructure Index (PIN)

Physical infrastructure index is based on five variables i.e. electricity production
(kWh), telephone lines (per thousand people), railway lines (thousand kilometers), total
road network (kilometers) and air transport freight (million tons kilometer). The expected
impact of physical infrastructure is indirect. It is assumed that poverty reduces as the
physical infrastructure in an economy develops.

Social Infrastructure Index (SIN)

Theoretically, it is observed that social infrastructure reduces poverty. The social
infrastructure index is comprised of health expenditures, education expenditures, life
expectancy and university enrolment.
Muhammad Zahir Faridi, Muhammad Omer Chaudhry, Muhammad Ramzan 539
Gross Domestic Product (GDP)
Gross Domestic Product at factor cost is used as proxy for national income. The
increasing trend in gross domestic product reduces poverty.

Inflation Rate (INF)

In recent past, inflation has become a very important factor for the stabilization of
the economy. We have used inflation rate to examine its impact on poverty. Empirically
and even theoretically, it has been noted that inflation has both supply side and demand
side effects.

Literacy Rate (LITR)

Education is considered to be a very important factor for poverty alleviation,
growth and employment generation. We have used literacy rate as proxy for education.
Empirically, it is observed that education escalates poverty.

V. Results and Discussions

This section interprets the results and findings of the study. Discussion about
findings is made in two ways. First we have discussed the description properties of data.
In the second stage, we have commented on the estimation analysis.

A. Descriptive Analysis
The descriptive statistics of some selected variables are given in table 6.

Table 6: Descriptive Analysis

Mean Median Maximum Minimum Std.Dev Skewness K J.B P
LFP 65.05 65.90 67.20 59.30 1.93 -1.23 3.97 11.38 0.0034
FRTY 5.49 5.60 7.00 3.40 1.29 -0.24 1.43 4.34 0.11
PIN 2.61 2.63 4.83 4.88 1.54 0.07 1.53 35.30 0.17
SIN 185081.4 62269.43 789267.6 12231.52 234086.1 1.40 3.53 13.38 0.00
GDP 5.15 4.80 9.00 1.70 1.88 0.12 2.14 1.28 0.52
LITR 39.88 39.60 59.00 19.73 13.15 0.03 1.50 3.64 0.16
INF 8.66 7.90 20.77 3.10 3.71 0.69 3.96 4.63 0.09
Source: Authors calculation

It is noted that on the average the labor force participation is 65.05 percent having
low standard deviation. The value of coefficient of skewness is negative. The value of
Jarque Bera is significant i.e. 11.38. Similarly the mean value of explanatory variables
are 5.49 for fertility rate, 2.61 for physical infrastructure index, 5.15 for GDP, 39.88 for
literacy rate and average inflation rate is 8.66 with standard deviation of
1.29,1.54,1.88,13.15 and 3.71 respectively.

We have reported the results of correlation among variables in table 7. The pair
wise correlation matrix shows the degree of associations among the variables. It is
observed that some variables are highly correlated while others are having low degree of
540 Pakistan Journal of Social Sciences Vol. 35, No. 2

Table 7: Correlation Matrix

FRTY -0.280 1
PI 0.190 -0.88 1
SI -0.320 -0.64 0.81 1
GDPG 0.031 0.21 -0.16 -0.11 1
LITR 0.190 -0.88 0.99 0.81 -0.19 1
INF -0.042 -0.15 0.14 0.26 0.08 0.09 1
Source: Authors calculations

B. Econometric Analysis
We have employed log log model specification to estimate the poverty model
and also to avoid the issue of Heteroskedasticity and model misspecification. The results
of the model are reported in table 8.

Table 8: Estimates of Poverty

Variable Coefficient Std. Error t-Statistic Prob.
C 36.313 3.895 9.321 0.0000
LNLFP -4.665 1.003 -4.648 0.0001
LNFRTY 0.311 0.156 1.993 0.0551
LNPI -0.920 0.211 -4.360 0.0001
LNSI -0.167 0.070 -2.377 0.0238
LNGDPG -0.047 0.046 -1.032 0.3096
LNLITR 2.703 0.581 4.652 0.0001
LNINF -0.037 0.043 -0.857 0.3978
2 2
R 0.811 Adjusted R 0.768
Durbin Watson 1.756 F statistic 19.006
Prob (F-statistic) 0.000
Source: Authors calculations

The values of coefficient determination and adjusted coefficient determination are

0.81 and 0.77 respectively. These indicate the explanatory power of the model is
powerful and significant because almost 77 percent variation in the poverty is explained
by the explanatory variables. Overall significance of the model is examined by F-
statistics which is highly significant at one percent level.

We have observed that in the present study there is no sign of multicollinearity

because adjusted R2 is less than 80 percent and almost all the variables are significant.
The problem of auto correlation is justified on the Durban Watson statistics. In our study,
the value of Durban Watson statistics is 1.76 which is near 2. It means that there is no
autocorrelation issue in the current analysis.

Labor force participation reduces the poverty. The value of the coefficient of labor
force participation is negative and statically significant. With one percent increase in
labor force participation; poverty level decreases about 4.7 percent. The reason may be
that when the working labor force increases, output and employment also increase that
become the cause of poverty reduction. Our findings stay in line with findings of
Kurosaki (2012).
Muhammad Zahir Faridi, Muhammad Omer Chaudhry, Muhammad Ramzan 541
We have found that fertility rate has positive and significant impact on poverty.
The co-efficient of fertility rate is positive. It shows if there is one percent increase in a
fertility rate 0.31 percent poverty will increase. The reason maybe that when more people
are added in the family and dependency burden increases and ultimately per capita
income decreases which is a sign of poverty.

The core variables of our study are physical infrastructure index and social
infrastructure index. We have observed in the study that physical infrastructure as well as
social infrastructure reduces the poverty. The co-efficient of both variables are negative
and statistically significant at one percent level. If there is one percent increase in
physical infrastructure index poverty decreases about 0.92 percent. Similarly if there is a
one percent increase in social infrastructure index poverty decline about 0.17 percent.
Findings are clearly explaining that physical infrastructure is powerful as compared to
social infrastructure index, though both reduce poverty. The reason may be that by
investing in social infrastructure as well as physical infrastructure growth rate of the
economy is accelerated, overall economic performance become better, employment
opportunities are created, income level of the people increases and become the cause of
poverty reduction.

Our study also includes gross domestic product, literacy and inflation rates as
explanatory variables. Gross domestic product and inflation reduces the poverty. The co-
efficient of both variables are negative and statically insignificant. The reason may be
that increasing growth rate is a sign of prosperity and well-being of the nation. The
negative influence of the inflation rate on poverty is justified because inflation increases
and unemployment decreases. According to the theory of Philips curve the increase in
money wage rate increases employment and poverty decreases. Finally we have noted
that literacy rate increases poverty. It is very interesting result. It indicates that educated
people are poor. The reason may be that highly educated people in developing countries
remain unemployed while searching for job that suits their qualification and standard.

VI. Conclusion
The results have been estimated by using OLS technique on time series data from
1975-2013. The results indicate that provision of social and physical infrastructures have
significant and positive impact on poverty alleviation. The study is unique in the sense
that the physical and social infrastructures index has been constructed by using different
variables. The results of this study have policy implications especially for developing
countries like Pakistan where the investment in infrastructure is a crucial decision given
the limited resources. It can be concluded that investment in infrastructure is a key
element in reducing poverty. It is, therefore suggested that government and policy makers
should make a feasible and non-discriminating policy to enhance the physical and social
infrastructure stock in the country in order to achieve the desired goals of economic
growth and reduction in poverty.

Ahmed, V., Abbas, A., and Ahmed, S. (2013). Public infrastructure and economic growth
in Pakistan: a dynamic CGE-micro simulation analysis Infrastructure and
Economic Growth in Asia in Cockburn, J., Dissou, Y., Duclos, J-Y. and Tiberti, L.
Ed Infrastructure and Economic Growth in Asia: 117-143 Springer.
542 Pakistan Journal of Social Sciences Vol. 35, No. 2

Aschauer D. (1989). Is public expenditure productive? Journal of Monetary Economics

Canning D, Pedroni P (2008). Infrastructure, long-run economic growth and causality
tests for co-integrated panels. Manchester School 76(5):504527
Devarajan S, Swaroop V, Zhou H (1996). The composition of public expenditure and
economic growth. Journal of Monetary Economics 37(3):313344
Faridi, M. Z., Malik, M. S., and Bashir, F. (2011). Transportation, Telecommunication
and Economic Development in Pakistan. Interdisciplinary Journal of Research in
Business 1(7): 45-52
Ghani E, Din M. (2006). The impact on public investment on economic growth in
Pakistan. Pakistan Development Review 45(1):8798
Imran, M., and Niazi, J. (2011). Infrastructure and Growth. The Pakistan Development
Review, 50(4-II): 355-364.
Jerome, A. (2011). Infrastructure, economic growth and poverty reduction in
Africa. Journal of Infrastructure Development, 3(2): 127-151.
Kurosaki, T. (2012). Urban transportation infrastructure and poverty reduction: Delhi
metros impact on the cycle rickshaw rental market: Institute of Economic
Research, Hitotsubashi University.
Ogun, T. P. (2010). Infrastructure and poverty reduction: implications for urban
development in Nigeria. Urban Forum. 21(3): 249-266
Pakistan Economic Survey, Ministry of Finance, Pakistan (various issues)
Planning Commission (2011). Framework for economic growth. Government of Pakistan,
Rehman JU, Iqbal A, Siddiqui MW (2010). Cointegration-causality analysis between
public expenditures and economic growth in Pakistan. European Journal of Social
Sciences 13(4):522
Sahoo, P., Dash, R. K., and Nataraj, G. (2010). Infrastructure development and economic
growth in China. IDE Discussion Paper # 261
Sala-i-Martin, X., Bilbao-Osorio, B., Blanke, J., Crotti, R., Hanouz, M. D., Geiger, T., &
Ko, C. (2012). The global competitiveness index 20122013: Strengthening
recovery by raising productivity. The Global Competitiveness Report 20122013
Sanchez-Robles B (1998). Infrastructure investment and growth: some empirical
evidence. Contemporary Economic Policy 16(1):98109
United Nations (2010). The Millennium Development Goals Report 2010.
World Bank (2006). World Development Report: Equity and Development. Oxford
University Press.